CYP2C19 Genotype Is Associated With Adverse Cardiovascular Outcomes in Black Patients Treated With Clopidogrel Undergoing Percutaneous Coronary Intervention

Abstract

Background: Cytochrome P450 2C19 (CYP2C19) intermediate and poor metabolizer patients exhibit diminished clopidogrel clinical effectiveness after percutaneous coronary intervention (PCI). However, outcome studies to date have lacked racial diversity. Thus, the impact of CYP2C19 genotype on cardiovascular outcomes in patients treated with clopidogrel who identify as Black or African American remains unclear.

Methods and results: Adults among 5 institutions who self-identified as Black or African American, underwent PCI and clinical CYP2C19 genotyping, and were treated with clopidogrel were included. Data were abstracted from health records. Major atherothrombotic (composite of death, myocardial infarction, ischemic stroke, stent thrombosis, or revascularization for unstable angina) and bleeding event rates within 1 year after PCI were compared across CYP2C19 metabolizer groups using multivariable Cox regression adjusted for potential confounders and baseline variables meeting a threshold of P<0.10. The population included 567 Black patients treated with clopidogrel (median age, 62 years; 46% women; 70% with an acute coronary syndrome indication for PCI). Major atherothrombotic events rates were significantly higher among clopidogrel-treated intermediate and poor metabolizers (24 of 125 [19.2%]) versus patients treated with clopidogrel without a no function allele (43 of 442 [9.7%]; 35.1 versus 15.9 events per 100 person-years; adjusted hazard ratio, 2.00 [95% CI, 1.20-3.33], P=0.008). Bleeding event rates were low overall (23 of 567 [4.1%]) and did not differ among the metabolizer groups.

Conclusions: Black patients with CYP2C19 intermediate and poor metabolizer phenotypes who are treated with clopidogrel exhibit increased risk of adverse cardiovascular outcomes after PCI in a real-world clinical setting. Bleeding outcomes should be interpreted cautiously. Prospective studies are needed to determine whether genotype-guided use of prasugrel or ticagrelor in intermediate and poor metabolizers improves outcomes in Black patients undergoing PCI.

Keywords: Black or African American; CYP2C19; clopidogrel; genetic testing; percutaneous coronary intervention; precision medicine.

Figure 1. Time‐to‐event curves for major atherothrombotic events (MAE) and major adverse cardiovascular events (MACE) among Black patients treated with clopidogrel by cytochrome P450 2C19 (CYP2C19) status.

Time‐to‐event curves for atherothrombotic events among Black patients treated with clopidogrel by CYP2C19 status. The unadjusted log‐rank P value for each Kaplan–Meier curve is presented. (A) MAE (primary end point) and (B) MACE (secondary end point) within 12 months after the index percutaneous coronary intervention (PCI). The Kaplan–Meier curve tails in Figure 1A were truncated at 360 days after the index PCI due to <10% of each stratum being available for follow‐up. In the normal metabolizer/rapid metabolizer/ultrarapid metabolizer (NM/RM/UM) group, 1 event occurred after day 360, during which time only 17 patients (3.8%) were still in follow‐up. No events occurred in the intermediate metabolizer/poor metabolizer (IM/PM) group after day 360 when only 4 (3.2%) patients were available for follow‐up.

Figure 2. Time‐to‐event curves for clinically significant bleeding events among Black patients treated with clopidogrel by cytochrome P450 2C19 phenotype status.

Time‐to‐event curves for clinically significant bleeding events among normal and increased function phenotypes. The unadjusted log‐rank P value for each Kaplan–Meier curve is presented. Clinically significant bleeding events are defined as moderate or severe/life‐threatening bleeding per the Global Use of Strategies to Open Occluded Arteries bleeding criteria within 12 months following the index percutaneous coronary intervention (PCI). (A) Normal metabolizer (NM) vs rapid metabolizer/ultrarapid metabolizer (RM/UM) phenotypes (primary analysis) and (B) NM vs RM and UM phenotypes (secondary analysis).